ON GASEOUS COMBUSTION. 481 



a speed of about 1,500 metres per minute, the explosion flame in the 

 case of the three mixtures (a) C 2 N 2 +0„ (b) C 2 N 2 +0 2 +N 2 , and 

 (c) G 2 N 2 +20 2 . In each case the flame was photographed, after detona- 

 tion had been set up, as it dashed past a glass window inserted 

 into the lead explosion coil. 1 The image obtained in the case of (a) 

 showed an intensely brilliant flame (the explosion wave), slightly drawn 

 out in tapering form ; in the case of (b), with nitrogen as diluent, the 

 flame was less brilliant and somewhat more drawn out than in (a) ; but 

 with (c) the flame, whilst no more luminous than in (b), was drawn out 

 to great length, owing to the continued combustion of carbon monoxide 

 in the rear of the wave. The following diagram, fig. 4 (approximately 

 to scale), will convey an idea of the relative durations of the flames in 

 the three cases : — 



Mixture C 2 N 2 + Z 



■25mm. * 



<95/TO77> 



/ b > iM^33iE3n=— Mixture C 2 N 2 i 2 + N 2 



» \ - - -40 mm - -- > 



Mixture 



(o n. — — -^~^-^^-' -• C 2 N 2 + 2 z 



«- - - - 130 mm. * 



Fig. 4. 



The Burning of Hydrocarbons in the Wave. — It was during the 

 course of measurements of the rates of explosion of hydrocarbons with 

 varying proportions of oxygen in 1891-92 that Dixon and his collabora- 

 tors rediscovered facts (originally set forth by Dalton in his ' New System 

 of Chemical Philosophy ') which finally disposed of the dogma that in a 

 deficient supply of oxygen a ' selective ' burning of hydrogen occurs. It 

 was found that when either ethylene or acetylene is detonated with its 

 own volume of oxygen the ultimate products consist almost entirely of 

 carbon monoxide and hydrogen. 2 The rates of explosion, whilst they 

 do not reveal the real mechanism of the combustion, show very clearly 

 that, as in the case of cyanogen, the carbon of a hydrocarbon is burnt 

 initially to the monoxide in the wave itself, the formation of the dioxide 

 being an after-effect and taking place in the rear of the wave. In the 

 cases of methane and acetylene the fastest rates are observed with 

 equimolecular mixtures, whereas with ethylene the rate increases with 

 the proportion of oxygen up to the limit C 2 H 4 +20 2 . As the question 

 of the mechanism of hydrocarbon combustion will be fully discussed 

 later, the following rates of explosion, as determined by Dixon, may 

 now be given without further comment : — ■ 



(A) Methane and Oxygen in Varying Proportions, fired at 10° and 760 mm. 



Mixture . CH 4 +0 2 CH 4 +1£0. 2 CH 4 +20 2 CH 4 + 30 2 * CH 4 +40 2 



Rate in metres \ 2)52g 2m 2>322 2UQ ]>963 



per sec. j 



* Compare this with the 2,154 metres per sec. observed for CH 4 +HOo-|-liNo 

 as showing the inertness of at least the moiety of the oxygen in the wave. 



1 Trans. Chem. Soc, 1890, 69, 759. 



' Vide Lean and Bone, Trans. Chem. Soc, 1892, 61, 873, and Bone and Cain, ibid., 

 1897, 71, 2G. 



